The Royal Australian and New Zealand College of Psychiatrists guidelines: Acute inpatient care for schizophrenia

The recent College Guidelines provide a comprehensive overview of the management of schizophrenia (Galletly et al., 2016), but their advice on acute inpatient care is necessarily limited because there is little research on the processes and outcomes of psychiatric inpatient treatment. Hence, there is minimal empirical information on either the risks or benefits of admission. For example, a Cochrane review located only six randomised controlled trials (RCTs) on the effects of inpatient length of stay (LOS) on the outcomes of severe mental illness, and all these studies were undertaken between 1969 and 1980 before the dramatic reductions in LOS (Babalola et al., 2014). This lack of research is surprising given that inpatient care has a unique role in the treatment of schizophrenia. While community and hospital-at-home teams can be effective for many (but not all) patients, inpatient care is essential when an acute episode is accompanied by potentially high risks of suicide or violence.

In the absence of an active Australian research programme on inpatient care, the default policy is minimising the costs of inpatient care by reducing psychiatric bed numbers and lowering the average LOS, partly through attempts at substitution with community sub-acute care. As the Guidelines note, Australian governments have already halved the funding for the inpatient care of schizophrenia (from an average annual cost per person of AUD22,715 in 1997–1998 down to AUD10,925 in 2010). What will happen to patient care if this trend continues into the next decade?

The Guidelines offer useful advice for psychiatrists if bed closures are increasing the pressure to reduce LOS by discharging patients too early. They provide Level I evidence for the length of time for an adequate trial of antipsychotic medication (at least 6 weeks of 300–1000 mg in chlorpromazine equivalents). The Guidelines’ advice is consistent with the RCT-level evidence for the time-course of improvement in acute schizophrenia with antipsychotics (Sherwood et al., 2005). Meta-analysis shows an early onset of action for antipsychotics with a gradual linear decline in total symptom scores from baseline to 4 weeks and a flattening of treatment response by 6 weeks.

Are Australian psychiatrists currently able to offer admissions that are long enough for adequate inpatient trials of antipsychotic medication, or are patients being discharged before achieving an antipsychotic response? Currently, Australian mental health patients have relatively short average LOS (16 days for all mental health patients and 24 days for schizophrenia in both public acute and non-acute hospital settings during 2012), which are below the Organisation for Economic Co-operation and Development (OECD) averages (28 days overall and 55 days for schizophrenia). However, even these Australian averages could be misleadingly high as most mental health patients have relatively short admissions (below the 2–4 weeks required for a clinical response to antipsychotic medication), with a relatively small proportion of patients receiving very extended admissions usually in long-term settings, which lifts the means (Victorian Department of Health and Human Services, 2016). Further detailed study is required of the Australian LOS for patients who are trialling new antipsychotic medications for an acute relapse of schizophrenia.

One concerning development for Australian psychiatrists is the national mental health policy presupposing further reductions in hospital LOS by transferring acute patients to community sub-acute settings. The recent evaluation of the large Prevention and Recovery Care (PARC) programme casts further doubt on this strategy. The PARC programme rolled out 230 sub-acute beds across Victorian metropolitan and rural communities to reduce acute hospital demand. It was found that schizophrenia spectrum disorders were indeed the most frequent diagnoses in both PARC services and inpatient wards. However, the evaluators concluded, ‘despite the introduction of PARC services, there have been no significant changes in use of or length of stay at inpatient mental health services’ from 2009 to 2014 (Victorian Department of Health and Human Services, 2016: 13). Essentially, PARC was unable to reduce the burden on Victoria’s acute hospitals and their emergency departments. This evaluation reinforces the impression that sub-acute care should not be viewed as a replacement for acute inpatient beds, but rather as an additional community service.

The United States offers a cautionary tale on the policy of minimising psychiatric bed numbers and shortening average LOS (Treatment Advocacy Center, 2016). The OECD indicated that the United States has even fewer psychiatric beds than Australia (25 vs 39 psychiatric beds per 100,000 population in both mental hospitals and acute hospitals compared to the OECD average of 68 psychiatric beds per 100,000 population), and managed care has reduced the average LOS to a mere 6 days (with 10 days for the inpatient treatment of schizophrenia). These ultra-short admissions can only offer quick assessment, sedation and acute stabilisation on the grounds of an immediate risk of suicide or violence. With very limited access to public psychiatric beds, many patients have been caught up in the nexus between care and punishment. There are high rates of schizophrenia among prisoners, and the few remaining US state mental hospital beds have virtually become an extension of the criminal justice system (Treatment Advocacy Center, 2016).

The American present could well be Australia’s future if funding for the inpatient care of schizophrenia is halved again over the next decade, and the average LOS is forced below the time required for an adequate biological response to antipsychotic medication. Given our responsibility as consultants for patients admitted to hospital for the acute treatment of schizophrenia, psychiatrists should lead the national debate on the safe minimum number of psychiatric beds for Australia and resist an unwarranted push towards the US system.

See Guideline by Galletly et al., 50: 410–472.